2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.datamodel.xdb.embl;
23 import jalview.analysis.SequenceIdMatcher;
24 import jalview.bin.Cache;
25 import jalview.datamodel.DBRefEntry;
26 import jalview.datamodel.DBRefSource;
27 import jalview.datamodel.FeatureProperties;
28 import jalview.datamodel.Mapping;
29 import jalview.datamodel.Sequence;
30 import jalview.datamodel.SequenceFeature;
31 import jalview.datamodel.SequenceI;
32 import jalview.util.DBRefUtils;
33 import jalview.util.DnaUtils;
34 import jalview.util.MapList;
35 import jalview.util.MappingUtils;
36 import jalview.util.StringUtils;
38 import java.text.ParseException;
39 import java.util.Arrays;
40 import java.util.Hashtable;
41 import java.util.List;
43 import java.util.Map.Entry;
44 import java.util.Vector;
45 import java.util.regex.Pattern;
48 * Data model for one entry returned from an EMBL query, as marshalled by a
52 * http://www.ebi.ac.uk/Tools/dbfetch/dbfetch?db=ena_sequence&id=J03321
55 * @see embl_mapping.xml
57 public class EmblEntry
59 private static final Pattern SPACE_PATTERN = Pattern.compile(" ");
65 String sequenceVersion;
73 String sequenceLength;
75 String taxonomicDivision;
79 String firstPublicDate;
81 String firstPublicRelease;
83 String lastUpdatedDate;
85 String lastUpdatedRelease;
87 Vector<String> keywords;
89 Vector<DBRefEntry> dbRefs;
91 Vector<EmblFeature> features;
93 EmblSequence sequence;
96 * @return the accession
98 public String getAccession()
105 * the accession to set
107 public void setAccession(String accession)
109 this.accession = accession;
115 public Vector<DBRefEntry> getDbRefs()
124 public void setDbRefs(Vector<DBRefEntry> dbRefs)
126 this.dbRefs = dbRefs;
130 * @return the features
132 public Vector<EmblFeature> getFeatures()
139 * the features to set
141 public void setFeatures(Vector<EmblFeature> features)
143 this.features = features;
147 * @return the keywords
149 public Vector<String> getKeywords()
156 * the keywords to set
158 public void setKeywords(Vector<String> keywords)
160 this.keywords = keywords;
164 * @return the sequence
166 public EmblSequence getSequence()
173 * the sequence to set
175 public void setSequence(EmblSequence sequence)
177 this.sequence = sequence;
181 * Recover annotated sequences from EMBL file
185 * a list of protein products found so far (to add to)
186 * @return dna dataset sequence with DBRefs and features
188 public SequenceI getSequence(String sourceDb, List<SequenceI> peptides)
190 SequenceI dna = new Sequence(sourceDb + "|" + accession,
191 sequence.getSequence());
192 dna.setDescription(description);
193 DBRefEntry retrievedref = new DBRefEntry(sourceDb,
194 getSequenceVersion(), accession);
195 dna.addDBRef(retrievedref);
196 // add map to indicate the sequence is a valid coordinate frame for the
198 retrievedref.setMap(new Mapping(null, new int[] { 1, dna.getLength() },
199 new int[] { 1, dna.getLength() }, 1, 1));
202 * transform EMBL Database refs to canonical form
206 for (DBRefEntry dbref : dbRefs)
208 dbref.setSource(DBRefUtils.getCanonicalName(dbref.getSource()));
215 for (EmblFeature feature : features)
217 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
219 parseCodingFeature(feature, sourceDb, dna, peptides);
222 } catch (Exception e)
224 System.err.println("EMBL Record Features parsing error!");
226 .println("Please report the following to help@jalview.org :");
227 System.err.println("EMBL Record " + accession);
228 System.err.println("Resulted in exception: " + e.getMessage());
229 e.printStackTrace(System.err);
236 * Extracts coding region and product from a CDS feature and properly decorate
237 * it with annotations.
242 * source database for the EMBLXML
244 * parent dna sequence for this record
246 * list of protein product sequences for Embl entry
248 void parseCodingFeature(EmblFeature feature, String sourceDb,
249 SequenceI dna, List<SequenceI> peptides)
251 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
253 int[] exon = getCdsRanges(feature);
258 Map<String, String> vals = new Hashtable<String, String>();
259 SequenceIdMatcher matcher = new SequenceIdMatcher(peptides);
262 * codon_start 1/2/3 in EMBL corresponds to phase 0/1/2 in CDS
263 * (phase is required for CDS features in GFF3 format)
268 * parse qualifiers, saving protein translation, protein id,
269 * codon start position, product (name), and 'other values'
271 if (feature.getQualifiers() != null)
273 for (Qualifier q : feature.getQualifiers())
275 String qname = q.getName();
276 if (qname.equals("translation"))
278 // remove all spaces (precompiled String.replaceAll(" ", ""))
279 prseq = SPACE_PATTERN.matcher(q.getValues()[0]).replaceAll("");
281 else if (qname.equals("protein_id"))
283 prid = q.getValues()[0];
285 else if (qname.equals("codon_start"))
289 codonStart = Integer.parseInt(q.getValues()[0]);
290 } catch (NumberFormatException e)
292 System.err.println("Invalid codon_start in XML for "
293 + accession + ": " + e.getMessage());
296 else if (qname.equals("product"))
298 // sometimes name is returned e.g. for V00488
299 prname = q.getValues()[0];
303 // throw anything else into the additional properties hash
304 String[] qvals = q.getValues();
307 String commaSeparated = StringUtils.arrayToSeparatorList(qvals,
309 vals.put(qname, commaSeparated);
315 DBRefEntry protEMBLCDS = null;
316 exon = MappingUtils.removeStartPositions(codonStart - 1, exon);
317 boolean noProteinDbref = true;
319 SequenceI product = null;
321 if (prseq != null && prname != null && prid != null)
324 * look for product in peptides list, if not found, add it
326 product = matcher.findIdMatch(prid);
329 product = new Sequence(prid, prseq, 1, prseq.length());
330 product.setDescription(((prname.length() == 0) ? "Protein Product from "
333 peptides.add(product);
334 matcher.add(product);
337 // we have everything - create the mapping and perhaps the protein
339 if (exon == null || exon.length == 0)
342 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
343 + sourceDb + ":" + getAccession() + ")");
344 if (prseq.length() * 3 == (1 - codonStart + dna.getSequence().length))
347 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
348 // this might occur for CDS sequences where no features are
350 exon = new int[] { dna.getStart() + (codonStart - 1),
352 map = new Mapping(product, exon, new int[] { 1, prseq.length() },
355 if ((prseq.length() + 1) * 3 == (1 - codonStart + dna.getSequence().length))
358 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
359 exon = new int[] { dna.getStart() + (codonStart - 1),
361 map = new Mapping(product, exon, new int[] { 1, prseq.length() },
367 // Trim the exon mapping if necessary - the given product may only be a
368 // fragment of a larger protein. (EMBL:AY043181 is an example)
372 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
374 // if given a dataset reference, search dataset for parent EMBL
375 // sequence if it exists and set its map
376 // make a new feature annotating the coding contig
380 // final product length truncation check
381 // TODO should from range include stop codon even if not in protein
382 // in order to include stop codon in CDS sequence (as done for
384 int[] cdsRanges = adjustForProteinLength(prseq.length(), exon);
385 map = new Mapping(product, cdsRanges, new int[] { 1,
386 prseq.length() }, 3, 1);
387 // reconstruct the EMBLCDS entry
388 // TODO: this is only necessary when there codon annotation is
389 // complete (I think JBPNote)
390 DBRefEntry pcdnaref = new DBRefEntry();
391 pcdnaref.setAccessionId(prid);
392 pcdnaref.setSource(DBRefSource.EMBLCDS);
393 pcdnaref.setVersion(getSequenceVersion()); // same as parent EMBL
395 MapList mp = new MapList(new int[] { 1, prseq.length() },
396 new int[] { 1 + (codonStart - 1),
397 (codonStart - 1) + 3 * prseq.length() }, 1, 3);
398 pcdnaref.setMap(new Mapping(mp));
401 product.addDBRef(pcdnaref);
402 protEMBLCDS = new DBRefEntry(pcdnaref);
403 protEMBLCDS.setSource(DBRefSource.EMBLCDSProduct);
404 product.addDBRef(protEMBLCDS);
408 // add cds feature to dna seq - this may include the stop codon
409 for (int xint = 0; exon != null && xint < exon.length; xint += 2)
411 SequenceFeature sf = makeCdsFeature(exon, xint, prname, prid, vals,
413 sf.setType(feature.getName()); // "CDS"
414 sf.setEnaLocation(feature.getLocation());
415 sf.setFeatureGroup(sourceDb);
416 dna.addSequenceFeature(sf);
421 * add dbRefs to sequence, and mappings for Uniprot xrefs
423 if (feature.dbRefs != null)
425 boolean mappingUsed = false;
426 for (DBRefEntry ref : feature.dbRefs)
429 * ensure UniProtKB/Swiss-Prot converted to UNIPROT
431 ref.setSource(DBRefUtils.getCanonicalName(ref.getSource()));
432 if (ref.getSource().equals(DBRefSource.UNIPROT))
434 String proteinSeqName = DBRefSource.UNIPROT + "|"
435 + ref.getAccessionId();
436 if (map != null && map.getTo() != null)
441 * two or more Uniprot xrefs for the same CDS -
442 * each needs a distinct Mapping (as to a different sequence)
444 map = new Mapping(map);
449 * try to locate the protein mapped to (possibly by a
450 * previous CDS feature)
452 SequenceI proteinSeq = matcher.findIdMatch(proteinSeqName);
453 if (proteinSeq == null)
455 proteinSeq = new Sequence(proteinSeqName,
456 product.getSequenceAsString());
457 matcher.add(proteinSeq);
458 peptides.add(proteinSeq);
460 map.setTo(proteinSeq);
461 map.getTo().addDBRef(
462 new DBRefEntry(ref.getSource(), ref.getVersion(), ref
466 noProteinDbref = false;
470 DBRefEntry pref = new DBRefEntry(ref.getSource(),
471 ref.getVersion(), ref.getAccessionId());
472 pref.setMap(null); // reference is direct
473 product.addDBRef(pref);
474 // Add converse mapping reference
477 Mapping pmap = new Mapping(dna, map.getMap().getInverse());
478 pref = new DBRefEntry(sourceDb, getSequenceVersion(),
479 this.getAccession());
481 if (map.getTo() != null)
483 map.getTo().addDBRef(pref);
489 if (noProteinDbref && product != null)
491 // add protein coding reference to dna sequence so xref matches
492 if (protEMBLCDS == null)
494 protEMBLCDS = new DBRefEntry();
495 protEMBLCDS.setAccessionId(prid);
496 protEMBLCDS.setSource(DBRefSource.EMBLCDSProduct);
497 protEMBLCDS.setVersion(getSequenceVersion());
499 .setMap(new Mapping(product, map.getMap().getInverse()));
501 product.addDBRef(protEMBLCDS);
503 // Add converse mapping reference
506 Mapping pmap = new Mapping(product, protEMBLCDS.getMap().getMap()
508 DBRefEntry ncMap = new DBRefEntry(protEMBLCDS);
510 if (map.getTo() != null)
520 * Helper method to construct a SequenceFeature for one cds range
523 * array of cds [start, end, ...] positions
524 * @param exonStartIndex
525 * offset into the exons array
527 * @param proteinAccessionId
529 * map of 'miscellaneous values' for feature
531 * codon start position for CDS (1/2/3, normally 1)
534 protected SequenceFeature makeCdsFeature(int[] exons, int exonStartIndex,
535 String proteinName, String proteinAccessionId,
536 Map<String, String> vals, int codonStart)
538 int exonNumber = exonStartIndex / 2 + 1;
539 SequenceFeature sf = new SequenceFeature();
540 sf.setBegin(Math.min(exons[exonStartIndex], exons[exonStartIndex + 1]));
541 sf.setEnd(Math.max(exons[exonStartIndex], exons[exonStartIndex + 1]));
542 sf.setDescription(String.format("Exon %d for protein '%s' EMBLCDS:%s",
543 exonNumber, proteinName, proteinAccessionId));
544 sf.setPhase(String.valueOf(codonStart - 1));
545 sf.setStrand(exons[exonStartIndex] <= exons[exonStartIndex + 1] ? "+"
547 sf.setValue(FeatureProperties.EXONPOS, exonNumber);
548 sf.setValue(FeatureProperties.EXONPRODUCT, proteinName);
551 StringBuilder sb = new StringBuilder();
552 boolean first = true;
553 for (Entry<String, String> val : vals.entrySet())
559 sb.append(val.getKey()).append("=").append(val.getValue());
561 sf.setValue(val.getKey(), val.getValue());
563 sf.setAttributes(sb.toString());
569 * Returns the CDS positions as a single array of [start, end, start, end...]
570 * positions. If on the reverse strand, these will be in descending order.
575 protected int[] getCdsRanges(EmblFeature feature)
577 if (feature.location == null)
584 List<int[]> ranges = DnaUtils.parseLocation(feature.location);
585 return listToArray(ranges);
586 } catch (ParseException e)
588 Cache.log.warn(String.format(
589 "Not parsing inexact CDS location %s in ENA %s",
590 feature.location, this.accession));
596 * Converts a list of [start, end] ranges to a single array of [start, end,
602 int[] listToArray(List<int[]> ranges)
604 int[] result = new int[ranges.size() * 2];
606 for (int[] range : ranges)
608 result[i++] = range[0];
609 result[i++] = range[1];
615 * truncate the last exon interval to the prlength'th codon
621 static int[] adjustForProteinLength(int prlength, int[] exon)
623 if (prlength <= 0 || exon == null)
627 int desiredCdsLength = prlength * 3;
628 int exonLength = MappingUtils.getLength(Arrays.asList(exon));
631 * assuming here exon might include stop codon in addition to protein codons
633 if (desiredCdsLength == exonLength
634 || desiredCdsLength == exonLength - 3)
642 origxon = new int[exon.length];
643 System.arraycopy(exon, 0, origxon, 0, exon.length);
645 for (int x = 0; x < exon.length; x += 2)
647 cdspos += Math.abs(exon[x + 1] - exon[x]) + 1;
648 if (desiredCdsLength <= cdspos)
650 // advanced beyond last codon.
652 if (desiredCdsLength != cdspos)
655 // .println("Truncating final exon interval on region by "
656 // + (cdspos - cdslength));
660 * shrink the final exon - reduce end position if forward
661 * strand, increase it if reverse
663 if (exon[x + 1] >= exon[x])
665 endxon = exon[x + 1] - cdspos + desiredCdsLength;
669 endxon = exon[x + 1] + cdspos - desiredCdsLength;
677 // and trim the exon interval set if necessary
678 int[] nxon = new int[sxpos + 2];
679 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
680 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon
687 public String getSequenceVersion()
689 return sequenceVersion;
692 public void setSequenceVersion(String sequenceVersion)
694 this.sequenceVersion = sequenceVersion;
697 public String getSequenceLength()
699 return sequenceLength;
702 public void setSequenceLength(String sequenceLength)
704 this.sequenceLength = sequenceLength;
707 public String getEntryVersion()
712 public void setEntryVersion(String entryVersion)
714 this.entryVersion = entryVersion;
717 public String getMoleculeType()
722 public void setMoleculeType(String moleculeType)
724 this.moleculeType = moleculeType;
727 public String getTopology()
732 public void setTopology(String topology)
734 this.topology = topology;
737 public String getTaxonomicDivision()
739 return taxonomicDivision;
742 public void setTaxonomicDivision(String taxonomicDivision)
744 this.taxonomicDivision = taxonomicDivision;
747 public String getDescription()
752 public void setDescription(String description)
754 this.description = description;
757 public String getFirstPublicDate()
759 return firstPublicDate;
762 public void setFirstPublicDate(String firstPublicDate)
764 this.firstPublicDate = firstPublicDate;
767 public String getFirstPublicRelease()
769 return firstPublicRelease;
772 public void setFirstPublicRelease(String firstPublicRelease)
774 this.firstPublicRelease = firstPublicRelease;
777 public String getLastUpdatedDate()
779 return lastUpdatedDate;
782 public void setLastUpdatedDate(String lastUpdatedDate)
784 this.lastUpdatedDate = lastUpdatedDate;
787 public String getLastUpdatedRelease()
789 return lastUpdatedRelease;
792 public void setLastUpdatedRelease(String lastUpdatedRelease)
794 this.lastUpdatedRelease = lastUpdatedRelease;
797 public String getDataClass()
802 public void setDataClass(String dataClass)
804 this.dataClass = dataClass;